Requesty
Tencent

minimax-m3

MiniMax-M3 is MiniMax's latest model, achieving frontier capabilities in professional tasks such as programming and agents. It uses the new attention architecture MSA (MiniMax Sparse Attention), supports up to 1M ultra-long context, and is a native multimodal model supporting image and video input.

VisionReasoningCaching
Try minimax-m3CompareDocs

Specifications

Context window1.0M tokens
Max output1.0M tokens
API typechat
AddedJun 23, 2026
Model IDtencent/minimax-m3
Data retentionYes
Used for trainingNo
Provider location🇨🇳 China
Privacy policyTencent Cloud Privacy Policy→

Benchmarks

Released 2026-06-01
Coding Indexcoding
58.6%

Artificial Analysis Coding Index — a composite of coding evaluations including LiveCodeBench, SciCode and Terminal-Bench.

GPQA Diamondreasoning
92.9%

Graduate-level physics, chemistry & biology questions designed to resist Googling.

Intelligence Indexreasoning
44.4%

Artificial Analysis Intelligence Index — a composite of multiple evaluations measuring overall model capability.

Scores are sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and do not capture every aspect of model quality. Always test on your own workload.

Pricing

Input / 1M
$0.30
Output / 1M
$1.20
Cache write
N/A
Cache read / 1M
$0.06
Estimated cost
100K input + 10K output$0.0420
1M input + 100K output$0.42
10M input + 1M output$4.20

Requesty charges exactly what the upstream provider charges, no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to tencent/minimax-m3.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="tencent/minimax-m3",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other Tencent models

deepseek-v4-pro
1.0M ctx$1.74/1M
Coding 59
deepseek-v4-flash
1.0M ctx$0.14/1M
Coding 56
kimi-k2.7-code
262K ctx$0.95/1M
Coding 61
glm-5.2
1.0M ctx$1.40/1M
Coding 69

Frequently asked questions

How much does minimax-m3 cost?
minimax-m3 is priced at $0.30 per million input tokens and $1.20 per million output tokens when accessed via Requesty. Prompt caching is supported, which can cut effective input cost by up to 90% on repeated context. Requesty charges exactly what the upstream provider charges, we don't add markup.
What is the context window of minimax-m3?
minimax-m3 has a context window of 1.0M tokens, with a maximum output of 1.0M tokens per response. That's roughly 1,398 words of input you can fit in a single prompt.
How does minimax-m3 perform on benchmarks?
minimax-m3 scores 92.9% on GPQA Diamond, 88.9% on τ²-Bench, 58.6% on Coding Index. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can minimax-m3 do?
minimax-m3 supports vision input, extended reasoning, prompt caching. You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use minimax-m3 with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "tencent/minimax-m3". The Quickstart above shows Python, JavaScript and cURL snippets.
Can I run minimax-m3 through Requesty?
Yes. minimax-m3 runs through Requesty's OpenAI-compatible API, served from Tencent. You do not host the model yourself: point base_url at Requesty, set the model to "tencent/minimax-m3", and requests are routed to the upstream provider with automatic failover. The same key gives you 400+ other models too.

Access minimax-m3 through Requesty

One API key, 400+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

Get started freeAll Tencent models